text note s1 a. geochronology and stratigraphy · the anterior inferior iliac spine (aiis) is...
TRANSCRIPT
1
Text Note S1 A. Geochronology and stratigraphy KSD-VP-1/1 was recovered from mudstone at the base of an exposure of upward coarsening claystone, siltstone and sandstone nearly 1 m thick (see Figure S2). Laterally, this upward coarsening succession is incised and overlain by channel-fill conglomerate and sandstone (Figure S2). A trench excavated about 25 m from the fossil locality reveals more than 2.5 m of claystone, siltstone and sandstone below the fossil horizon (see Figure S3). A tuff at the base of the trench has been dated by the 40Ar/39Ar method, yielding an age of 3.60 ± 0.03 Ma (2; see Figure S4 and Table S2)1. All strata in this section comprise a single normal magnetozone (see below for details). Samples with previously reported reverse magnetization2 are from an adjacent fault block with older stratigraphic units (see Figure S2). The simplest interpretation of the age and reversal data is that the bottom of the trench lies at or just above the Gauss/Gilbert paleomagnetic transition, with the normal magnetozone equating with the lower part of the normal subchron C2An.3n (3.596–3.330 Ma3). Using nominal sedimentation rates (11 cm/ky overall for strata ~3.8–3.6 Ma in this area1), the fossil specimen KSD-VP-1/1, found 2.6 m above the dated tuff, has an estimated age of 3.58 Ma. B. Paleomagnetic samples and results Nine oriented blocks for paleomagnetic analysis (KSD09-1 to KSD09-9) were collected from claystone and siltstone strata in a 3m deep trench 25 m from KSD-VP-1/1 partial skeleton at 11º28’54.5’’N, 40º30’39.8’’E. Three specimens were cut from each block and the natural remanent magnetization (NRM) measured at the Berkeley Geochronology Center, using alternating field (AF) followed by thermal demagnetization at ~50ºC steps to 605ºC (for further discussion of technique see 1). All samples exhibit normal remanence directions, though some (KSD09-1, 09-3 and 09-7) show a reverse overprint typically removed during AF and thermal demagnetization (Figures S4-S5). In a few specimens, the secondary reverse magnetization was not completely removed during the demagnetization procedure, leading to ambiguous polarity determinations. In these cases, new specimens were cut from the original sample blocks, selecting the finest grained material. For KSD09-1, the new specimens yielded unambiguous normal polarity. KSD09-7 showed normal declinations but shallow negative inclinations interpreted as a later, secondary overprint. Some specimens of KSD09-3 showed initial reverse directions that shifted to normal during the demagnetization process, with just a single specimen retaining reversed polarity after demagnetization. We conclude that all sedimentary rocks in the trench were deposited during a period of normal polarity (Figure S5).
2
Text Note S2 The anterior inferior iliac spine (AIIS) is massive and defines the upper border of a
deep iliopsoas groove (Figure S6). The ilium is thickened posterior to it, and while much of its crest is missing, the preserved portion (i.e., the "iliac pillar") is located more anteriorly than its human homologue, consistent with lateral iliac flare typical of Australopithecus4 .
The greater sciatic notch is well-demarcated and narrow, and is much more acute than that of A.L. 288-1 or Sts 14. The notch is substantially obtuse in two obvious female Australopithecus, but narrowed and more acute in KSD-VP-1/1. A platypelloid pelvis would not prevent selection for anteroposterior expansion of the birth canal. Was selection for AP birth canal already present in Australopithecus? The shape of the sciatic notch in the recently described BSN49/P27 pelvis, assigned to H. erectus, is relevant in this regard (see Figure S9). The ischial spine of KSD-VP-1/1 has been sheared away and a comparable value must therefore be estimated, but must have been between 0.20 and 0.25, below the human female range (ca 0.28-0.53) and well within the modern male range (ca 0.08-0.43). A.L. 288-1 and Sts 14 are 0.29 and 0.38, respectively. The former value lies within the range of overlap of human males and females, while Sts 14's value is well above the modern male range. The acetabulum has a well-formed lunate surface with a more substantial anterior extension than in A.L. 288-1. A portion of the superior pubic ramus is preserved, but has been compressed inferiorly, artificially reducing the acetabulum's diameter. Consistent with hip mechanics in Australopithecus4, an estimate of the original acetabulum appears somewhat small (vertical acetabular diameter ≈ 49 mm); i.e., it is about 14% of tibial length (see also below). Text Note S3
Ape distal tibial metaphyses are demonstrably broad and rectangular, primarily reflecting their more massive medial malleoli and expanded anterior plafonds5,6. Overall metaphyseal dimensions are unaffected by the specimen's pathology. Its AP/ML ratio (100) is decidedly hominid as is its probable malleolar index (= 47)(Table S7). Simple visual inspection shows the plafond to be remarkably narrow and largely square, rather than rectangular as in African apes (Figure S12). All of these features confirm a lack of any significant adaptation to vertical climbing5,6, as is the case with all early hominids.
KSD-VP-1/1e exhibits only one non-hominid character, a lateral declination of the distal articular surface (Table S7). Whereas the plafond/shaft angle is 90° in humans5, it is 100° in KSD-VP-1/1e. However, thirteen other early hominids have human-like values5, and fibular fracture (see earlier) is known to lead to lateral declination of the plafond7 as in KSD-VP-1/1e.
3
Figure S1. Satellite imagery location map of the Woranso-Mille study area. Korsi Dora is the locality where the partial skeleton (KSD-VP-1/1) was found. Other fossil collection areas are indicated.
4
Figure S2. Stratigraphic columns showing the positions of samples for which paleomagnetic data were previously published1. Solid circles indicate samples with reverse magnetization, all of which lie below the Kilaytoli Tuff (KT). Open circles indicate samples with normal magnetization, all from a different fault block than the reverse magnetized samples, and stratigraphically above the KT. Samples from both fault blocks were previously plotted, mistakenly, on a single composite section1.
5
Figure S3. Paleomagnetic results for nine samples collected along the excavated trench 25 m from the fossil site. Paleomagnetic directions show the angular distance (delta) to the expected normal direction (000º, 22) for this latitude. The stratigraphic position of the fossil skeleton horizon is indicated by the bone symbol. All samples exhibit normal remanence directions, with mean declinations towards the magnetic north, though some specimens in samples (KSD09-1, 09-3 and 09-7) show a reverse overprint producing shallow negative inclinations with a small delta value. A tuff at the base of this sequence has been dated by the 40Ar/39Ar method, yielding an age of 3.60 ± 0.03 Ma and all strata in this 3m section comprise a single normal magnetozone. The simplest interpretation is that the bottom of the trench lies at or just above the Gauss/Gilbert paleomagnetic transition, with the normal magnetozone equating with the lower part of the normal subchron C2An.3n (3.596–3.330 Mtea2). Using nominal sedimentation rates (11 cm/ky overall for strata ~3.8–3.6 Ma in this area1), the fossil specimen KSD-VP-1/1, found 2.8 m above the dated tuff, has an estimated age of 3.58 Ma.
6
Figure S4. Age-Probability Spectrum for Sample WM09/KSD-1 (Lab ID#s 25333, 25334).
7
Figure S5. Orthogonal and stereographic plots showing directions at different steps of demagnetization for specimens KSD09-3d and KSD09-8c. Removal of a reverse directed magnetization occurs in specimen KSD09-3d showing initial reverse directions that change to normal during demagnetization. Specimen KSD09-8c shows only normal directions.
8
Figure S6. Anteroposterior X-ray of KSD-VP-1/1d (left) and AP View of the Specimen (right). The radiograph was taken prior to partial repositioning of some fragments in the laboratory following removal of matrix, and the orientation of these two illustrations varies slightly. Note the broken and translated superior pubic ramus (white arrow on left) and the somewhat ovoid appearance of the acetabulum. Its elongated mediolateral transect is a consequence of superoinferior compression--the same damage that has caused the fragmentation of the superior pubic ramus. Note also the translation of a substantial portion of the retroauricular region (white arrow on the right). Although this fragment has been relocated since the X-ray, it nevertheless still lies considerably superior to its original location, and has resulted in superior repositioning of a major portion of the retroauricular region (including the PSIS) even after this repair. This has elevated the posterior iliac crest to some extent. For discussion see text.
9
Figure S7. Superior (left) and Anterior Views (right) of KSD-VP-1/1t. The alae are unusually broad as in other sacra of Australopithecus (cf., Figure S10).
10
Figure S8. Relative Ischial Length in KSD-VP-1/1d and Extant Hominoids. The minimum distance between the lower rim of the acetabulum and the superomost point on the ischial tuberosity has been divided by the acetabular diameter for samples of extant hominoids (N = 25 each species). The positions of the three Australopithecus specimens are close to those of Homo, but still lie slightly outside the range of the sample shown here, although this ratio in other human samples sometimes overlaps those of Australopithecus. At least some of this difference is a consequence of expansion of the diameter of the acetabulum in Homo. Values for Ar. ramidus (ARA-VP-6/500) are from the original specimen, but normalized by estimates of the probable maximum and minimum of acetabulum diameter. Each box shows the median, interquartile range (the box length), outliers (open circles) and extreme values (asterisks) within a category. Outliers are cases with values between 1.5 and 3 box lengths from the upper or lower edge of the box. Extreme values are those with values more than 3 box lengths from the upper or lower edge of the box.
KSD-VP-1/1d
A.L.-288-1
STS-14
ARA-VP-6/500
Homo
Gorilla
Pan
1.20
1.00
0.80
0.60
0.40
0.20
0.00
Rel
ativ
e Is
chia
l Len
gth
11
Figure S9. Plot of Estimated Angulation of Sciatic Notch. The shape of the greater sciatic notch is shown here graphically using the method and data of Simpson et al.8. That sample is composed of 20 females and 20 males from the Hamann-Todd collection, Cleveland Museum of Natural History. The horizontal axis is the range of values of the position of the deepest part of the notch relative to the "sacro-iliac tubercle along a line connecting the ischial spine to the tubercle for the inferior sacro-iliac ligament. A value of 0.5 indicates that the inflection point is located midway along a line defining the opening of the [greater sciatic notch]". A range of possible values is shown for KSD-VP-1/1 (connected by horizontal line). Values for A.L. 288-1, Sts-14, and BSN49/P27 are also shown. Body size suggests that KSD-VP-1/1 is male, while it is reasonably certain that A.L. 288-1 is female based on the same criterion (see also 9). These data may require reassessment of the assumption that pelvic dimorphism is not detectable in early hominids on the basis of pelvic morphological characters. Figure courtesy of S. Simpson.
KSD-VP-1/1
A.L. 288-1 Sts 14 BSN49/P27
12
Figure S10. Natural Log-Log Scatterplot of Relative Alar Breadth (both alae) of the Sacrum in Australopithecus and Extant Hominoids. Alar breadth was calculated as the maximum breadth of the sacrum less the transverse diameter of the S1 centrum. S1 centrum area was calculated as the product of its AP and ML dimensions. The human mean has also been plotted. Other symbols are provided in the legend. KSD-VP-1/1 lies nearer the human cloud than do A.L. 288-1 and Sts-14, which may be an allometric effect of these smaller examples of Australopithecus. Data for Sts-14 from Robinson10. Data for BSN49/P27 from the Simpson, et al. reconstruction8.
4.25
4.00
3.75
3.50
3.25
3.00
Lo
g A
lar
Bre
adth
7.757.507.257.006.756.506.256.00
Log S1 Centrum Area
Human Mean
STS-14
KSD-1/1
BSN49/P27
A.L. 288-1
Gorilla
Pan
Homo
13
Figure S11. Lateral (A) and Distal (B) Views of KSD-VP-1/1c. Both views are reproduced here in Black & White to highlight surface contrast. In A, material has been added between various cracks in order to stabilize the specimen, but the specimen's original dimensions were preserved by underlying cancellous bone. The condylar outline is decidedly elliptical, although the shape index11 falls approximately between human and chimpanzee distributions (see also text). B. The medial condyle is somewhat foreshortened by postmortem damage but its relationship to the lateral condyle is reasonably well preserved. A dashed line has been inscribed connecting the locations of the medial and lateral meniscal grooves. The lateral condyle's meniscal groove is difficult to locate because of surface damage, but the inscribed line passes through a distinct depression, which although eroded, was its probable location. Tibial position in full extension, as defined by this line (see also 11), shows a very high lateral lip to the patellar groove, suggesting a long period of substantial selection for a knee joint habitually placed in valgus during stance phase.
14
Figure S12. A. Lateral (A) and Inferior (B) Views of Distal End of KSD-VP-1/1e. Photographs A and B are reproduced here in Black & White to highlight surface contrast. A. A continuous perimeter of reactive bone surrounds the fibular facet. Its superior course has been marked by black arrows. The bony outgrowth has elongated tibial length by about one cm. This is most likely due to a remote, non-united fibular fracture. B. Inferior view shows a distal projection of the bony outgrowth (bracket) and the severed medial malleolus (arrow). The form of the malleolus, however, can be deduced because of its near-surgical transection. Despite the effects of the fracture, the markedly concave tibial plafond is still evident in A, though metric assessment requires caution because of the efflorescence of reactive bone. For further discussion see text.
15
Figure S13. KSD-VP-1/1e is nearly complete for its entire length (from well above the apex of the tibial tuberosity [arrow] proximally to the plafond surface distally), making length estimation very accurate, as only the intercondylar eminence is missing from some standard length measures, i.e., physiological length. The shaft has spalled and been compressed mediolaterally, but was nevertheless clearly platycnemic. Maximum Length = 355; Physiological Length = 343.For discussion see text.
A
16
Figure S14. Ribs Recovered from KSD-VP-1/1. Close-ups and drawings of the second rib are provided in Figures S15-S17. Specimens shown here (top to bottom) are KSD-VP-1/1n (left second rib; reversed), KSD-VP-1/1q (right 5th, 6th, or 7th rib), KSD-VP-1/1s (middle rib fragment), KSD-VP-1/1o (right 7th or 8th rib), KSD-VP-1/1p (right 8th or 9th rib), and KSD-VP-1/1r (left 11th rib). See also, Table S1.
17
Figure S15. Inferior (left) and Superior (right) Views of Left Second Rib. The rib's geometry is well preserved and its head also preserves its fundamental morphology. Slight postmortem damage at the point of reinforcement shown in both views may have slightly increased the rib's curvature (and thereby decreased estimates of original cupola volume). The specimen, however, is sufficiently complete to allow reliable assessment of curvature as shown in Figure S16-S17. Results of this assessment (Figure 2) place its curvature well within the human range and substantially below values of this index for most African apes. For additional discussion see the legends of Figure S16-S17 and the text.
18
Figure S16. Second Ribs of Hominoids Aligned on their Costal Neck Axis and Size Normalized. The second ribs of a chimpanzee (left), KSD-VP-1/1 (center), and a modern human (right) have been aligned on their costal neck axis12, and scaled to the same interfacet distance (distance A in Figure S17). Although the specimen from KSD-VP-1/1 is incomplete, enough of its original length was recovered to preserve the likely trajectory of its overall curvature, which is not constricted like that of Pan, and is therefore more similar to that of Homo. This can be quantitatively assessed using the index described in Figure S17. Results of that method are presented in Figure 2. The human and chimpanzee specimens have been redrawn from 12.
19
Figure S17. Method Used for Assessing Second Rib Curvature Index in Hominoids (cf., Figures 2 and S14-S16). A line is first drawn connecting the centroids of the two articular facets of the rib (for the centrum and for the transverse process of its associated vertebra); the "costal neck axis" (see 12 for more details). A line is then inscribed perpendicular to that line and which also lies tangent to the most lateral point on the pleural surface of the rib. Two distances are measured: one between the centroids of the two facets (A) and one between the centrum's facet and the intersection of the inscribed perpendicular with the costal neck axis (B). The Rib Curvature Index is then calculated as A/B X 100. Data for hominoids are shown in Figure 2 (Homo N = 13; Gorilla N = 33; Pan N = 16). The African ape and human ranges overlap only minimally; i.e., only one Gorilla value (65) fell within the human range, all others exceeded it. The value for KSD-VP-1/1n was 66.
20
Figure S18. Superior (above) and Inferior (below) Views of the Left Clavicle
(KSD-VP-1/1f). Both sternal and acromial ends have suffered bone loss (for discussion see text). Comparisons of regressed clavicle length to various individual upper and lower limb dimensions (Figure S19) show that KSD-VP-1/1f falls consistently with Homo and not with either African ape. In conjunction with the specimen's second rib, these data confirm that early hominids lacked the constricted cupola of African apes.
21
Figure S19. Box plots of Relative Clavicle Circumference and Length. Clavicular circumference (left) and length (right) have been divided by the geometric mean of eight upper limb joint dimensions (see Figures S28-S30). Note that relative clavicular circumference in KSD-VP-1/1 exceeds the range in African apes and falls near the upper limit of Homo. The latter is not surprising as the human antebrachium has undergone significant reduction relative to its length since BOU-VP-12/113. Regressed clavicle length lies closer to the human median; however, comparison of this length estimate (156 mm) to the actual specimen suggests that it may be an underestimate. Exfoliation at the midshaft of KSD-VP-1/1f also makes the regressed length an underestimate. Upper limb joint dimensions included in the geometric mean: superoinferior and maximum mediolateral dimensions of the scapular glenoid, articular breadth of the trochlea and capitulum and breadth of the trochlea, and the proximodistal height of the trochlear notch (taken from anterior-most points of the notch), the maximum depth of the trochlea taken perpendicular to the previous measurement, and the anteroposterior length of the coronoid process. For definition of the ulnar metrics, see 14, especially page 303 (see also 15). The radial facet of the ulna of KSD-VP-1/1 is too damaged for accurate measurement and was not included here. For box plot symbol parameters see the legend of Figure S8.
KSD-VP-1/1HomoGorillaPan
1.80
1.60
1.40
1.20
1.00
0.80Cla
vicl
e C
irc
um
fere
nce
Div
ided
by
Geo
me
tric
Me
an o
f U
pp
er L
imb
KSD-VP-1/1HomoGorillaPan
6.00
5.50
5.00
4.50
4.00
3.50
3.00Cla
vic
le L
eng
th D
ivid
ed b
y U
pp
er
Lim
b G
eo
met
ric
Mea
n
22
Figure S20. Log-log Scatterplot of Clavicle Length and the Geometric Mean of Eight Joint Dimensions of Scapula, Humerus, and Ulna. KSD-VP-1/1f is most similar to humans, although relative clavicle length does not differ significantly in Pan and Homo, and the distributions of the two taxa overlap. The 80% confidence limits for the KSD-VP-1/1 regressed length are shown. These suggest more gracile upper limb joints or longer clavicles in early hominids compared to apes, or both, but neither possibility is consistent with a pyramidal cupola. See Figures S28-S30 for variables included in the forelimb geometric mean.
23
Figure S21. Scapular Fragments of A.L. 288-1 (left; cast) and an Amerindian of Approximately Equal Size and Preservation. The specimens are from opposite sides and have been aligned based on the orientations of their glenoid planes. Note the virtual identity of their bar-glenoid angles.
24
Figure S22. Natural Log-Log Scatterplot of the Areas of the Supraspinous and Infraspinous Fossae from Extant Hominoids. Homo exhibits a smaller supraspinous area relative to its infraspinous area than do Pan and Gorilla. Although insufficient preservation prevents calculation of this ratio in KSD-VP-1/1g, it was likely more similar to those of Pan and Gorilla based on data from DIK-1-1. However, the dimensions reported for that specimen were simply linear distances. Areas shown here were calculated from X-rays using Image-J. A line was extended from the intersection of the scapular spine and vertebral border of the scapula to the base of the spine near its junction with the glenoid as seen on each X-ray. Each X-ray was made with the scapular blade elevated above the X-ray plate just sufficient to make the glenoid appear as a single broad curved line (cf., Figures 3 and S23). Once the "spine line" was defined and inscribed, the areas of the supraspinous and infraspinous fossae were calculated using the perimeter of the scapular blade as their remaining borders. These same X-rays were also used to calculate the various angles reported in the text and in Table 1.
10.009.509.008.50
Log Infraspinatus Area
10.00
9.50
9.00
8.50
8.00
7.50
7.00
6.50
Lo
g S
up
rasp
inat
us
Are
a
Gorilla
Pan
Homo
25
Figure S23. Definitions of Angles Used to Evaluate Geometry of Hominoid Scapulas. A human scapula is shown here with the five angles used in the PCA described in the text. CD-EF = axillary-vertebral angle; GH-AB = glenoid-spinal angle; AB-IJ = bar-glenoid angle; AB-CD = glenoid-axillary angle; GH-CD = spinal-axillary angle. For discussion see text.
26
Figure S24. Bar-Glenoid Angle in KSD-VP-1/1g, and in A.L. 288-1, Pan, Gorilla, and Homo. The human sample does not contain the specimen shown in Figure S21. For Box plot parameters see Figure S8. For definitions of angles see Figure S23.
A.L.288-1KSD-VP-1/1Gorilla Pan Homo
140.00
120.00
100.00
Bar
Gle
no
id A
ng
le
27
Figure S25. Axillary-Spine Angle in KSD-VP-1/1g, and in Pan, Gorilla, and Homo. For Box plot symbols see Figure S8. For definitions of angles see Figure S23.
KSD-VP-1/1Gorilla Pan Homo
60
40
20
Axi
llary
-sp
ine
An
gle
28
Figure S26. Anterior Views of Right Ulna (KSD-VP-1/1a [left]) and Right Humerus (KSD-VP-1/1b [right]). Both specimens have suffered considerable exfoliation of bone from their shaft surfaces. However, the deltopectoral crest of the humerus is clearly rugose. For discussion see text.
29
Figure S27. Calculation of Trochlear Notch Orientation (TNO) in Incomplete Ulnas (right) and Results for Hominoids and Early Hominids (below). The method used in the present study to measure TNO generally follows that used by Drapeau14 and Drapeau et al.15 for complete ulnas. Shown here is a measure devised to produce a similar metric from an incomplete ulna16. A photograph was taken of the lateral surface of each ulna. The distance from the most proximal point on the olecranon to an anteroposterior tangent applied to the distal edge of the radial facet was measured and doubled. Each photograph was then distally transected at that distance. A best-fit tangent was then applied to the posterior shaft as imaged here (following truncation) and a line connecting the most anterior projections of the coronoid and olecranon portions of the trochlear notch was then inscribed. Angle alpha was calculated and provided in Table S10. The results from this method largely parallel those obtained in 15, although the values listed in Table S10 reported here are higher than those reported in 15 and 16, because shaft curvature tends to reduce the value of the angle. The TNOs of early hominids differ significantly from those of African apes and modern humans, suggesting that an anterior-facing trochlear notch was unique to early hominids. For discussion see text.
30
Figure S28. Box plots of Square Root of Superoinferior and Mediolateral Dimensions of the Glenoid Compared to Tibia Length in KSD-VP-1/1 and A.L. 288-1. Tibia length in A.L. 288-1 was calculated using the crural index which varies from 80 to 85 in extant hominoids (cf., Table S12). For box plot parameters see Figure S8. For discussion see text.
A.L. 288-1KSD-VP-1/1HomoGorillaPan
0.15
0.125
0.10
0.075
Sq
uar
e R
oo
t G
len
oid
SI X
ML
Div
ided
by
Tib
ia L
eng
th
31
Figure S29. Box plots of Geometric Mean of Proximal Ulnar Metrics Compared to Tibia Length in KSD-VP-1/1 and A.L. 288-1. Tibia length in A.L. 288-1 was calculated using the crural index which varies from 80 to 85 in extant hominoids (cf., Table S12). For box plot parameters see Figure S8. Metrics used for the ulnar geometric mean are the proximodistal height of the trochlear notch, the maximum depth of the trochlea perpendicular to the previous metric, the anteroposterior length of the coronoid process, and the proximodistal height of the olecranon. Each of these is defined in 14 and 15. For discussion see text.
A.L. 288-1KSD-VP-1/1HomoGorillaPan
0.11
0.10
0.09
0.08
0.07
0.06
0.05
0.04Geo
met
ric
Mea
n P
roxi
mal
Uln
a D
ivid
ed b
y T
ibia
Len
gth
32
Figure S30. Box plots of Geometric Mean of Distal Humeral Metrics Compared to Tibia Length in KSD-VP-1/1 and A.L. 288-1. Tibia length in A.L. 288-1 was calculated using the crural index which varies from 80 to 85 in extant hominoids (cf., Table S12). For box plot parameters see Figure S8. Metrics used are the maximum transverse breadth of the distal humerus (medial to lateral epicondyles), and mediolateral breadth of the trochlea. For discussion see text.
A.L. 288-1KSD-VP-1/1HomoGorillaPan
0.20
0.18
0.16
0.14
0.12
0.10
0.08
Geo
met
ric
Me
an D
ista
l Hu
me
rus
Div
ide
d b
y T
ibia
Len
gth
33
Table S1
Elements Recovered from the KSD-VP-1/1 Partial Skeleton Assigned Accession Number
Element
KSD-VP-1/1a right ulna KSD-VP-1/1b right humerus KSD-VP-1/1c left distal femur KSD-VP-1/1d right innominate KSD-VP-1/1e left tibia KSD-VP-1/1f left clavicle KSD-VP-1/1g right scapula KSD-VP-1/1h cervical vertebra KSD-VP-1/1i cervical vertebra KSD-VP-1/1j cervical vertebra KSD-VP-1/1k cervical vertebra KSD-VP-1/1l cervical vertebra KSD-VP-1/1m vertebral body KSD-VP-1/1n left second rib KSD-VP-1/1o right lower rib (7th or 8th) KSD-VP-1/1p right lower rib (8th or 9th) KSD-VP-1/1q right upper rib (5th, 6th or 7th) KSD-VP-1/1r left 11th rib KSD-VP-1/1s middle rib fragment KSD-VP-1/1t superior sacral body and ala KSD-VP-1/1u Posterior sacral spine fragment KSD-VP-1/1v Sacral spine fragment KSD-VP-1/1w coccygeal body? KSD-VP-1/1x R. Spinous process with Sup. and Inf. facets KSD-VP-1/1y R. Spinous process and Sup. with Inf. facets KSD-VP-1/1z R. Spinous process and Sup. with Inf. facets KSD-VP-1/1aa L. Spinous process and Sup. with Inf. facets KSD-VP-1/1ab L. Spinous process and Sup. with Inf. facets KSD-VP-1/1ac L. Spinous process and Sup. with Inf. facets KSD-VP-1/1ad vertebral body fragment KSD-Vp-1/1ae left superior facet of vertebra KSD-VP-1/1af left superior facet of vertebra
34
Table S2
35
Table S3 Cardinal Characters of Preserved Portion of the KSD-VP-1/1d Os Coxa Compared to Those of Other Hominoids
Character / Taxon
Pan Ardipithecus ramidus
KSD-VP-1/1d A.L. 288-1 Homo
Size and form of Ilium
Narrow superoinferiorly elongated, with narrow g. minimus origin along anterior edge
Substantial anteroposterior expansion including origin of lesser gluteals
Dramatic anteroposterior expansion with extensive origin for lesser gluteals
As in KSD-VP-1/1 As in KSD-VP-1/1 but somewhat anteroposteriorly shorter because of change in birth canal architecture
Size and form of Ischial Tuberosity
Massive, with arc-like undifferentiated hamstring/adductor surface
As in Pan Greatly reduced with marked angulation of hamstring/adductor surface
As in KSD-VP-1/1 As in KSD-VP-1/1 but usually with slightly narrower gap between acetabulum and ischial tuberosity
Lesser sciatic notch
Elongate and only marginally defined
Moderately abbreviated distance between ischial spine and auricular surface with obtuse angulation
Greatly abbreviated distance between ischial spine and auricular surface with narrow acute angulation
As in KSD-VP-1/1 but with less acute angle of notch
Varies with sex. Most Homo males as with KSD-VP-1/1; most females as with A.L. 288-1
Iliac fossa Narrow, with almost no curvature when viewed from dorsal aspect
Broad but with only minimal sigmoid curvature when viewed from dorsal aspect
Extremely broad with sigmoid curvature when viewed from dorsal aspect
As in KSD-VP-1/1 Less broad than in Australopithecus and with more pronounced sigmoid curvature
Sacrum Very narrow overall and with narrow alae
None preserved Very broad, with high alae/sacral centrum area ratio
As in KSD-VP-1/1 Broad as in Australopithecus but with reduced alae/centrum ratio
36
Obturator internus groove
None Locus not preserved but present in likely ancestral forms (Orriron tugenensis)
Deep and well defined
As in KSD-VP-1/1 As in KSD-VP-1/1
Iliopsoas groove None Moderate in breadth and depth
Deep, well defined, with steep sides
As in KSD-VP-1/1 As in KSD-VP-1/1
AIIS by separate ossification center
No Adult, but almost certainly
Adult, but almost certainly
Adult, but almost certainly (present in Au. africanus)
Yes
Form of Anterior inferior iliac spine (AIIS)
Limited to rugose enthesis with only slight to moderate elevation above iliac surface
Large, sigmoid shaped, projecting
Large, sigmoid shaped, projecting
As in KSD-VP-1/1 As in KSD-VP-1/1
Iliac pillar None--Expansion only at anterior edge of ilium
Broadly expanded anterior iliac thickness
Anteriorly located and distinct
As in KSD-VP-1/1 Pillar well defined but located more posteriorly
Iliopubic eminence
None Damaged Damaged, but highly probable
No Typical and very frequent
Lunate surface of acetabulum
Covers most of acetabulum except acetabular notch
Damaged Covers all of acetabulum except acetabular notch
As in KSD-VP-1/1 As in KSD-VP-1/1
Auricular Surface
Highly elongate and very narrow
Damaged Narrow and moderately elongate
Broader and less elongate; more Homo-like
Broad and reniform with distinct superior and inferior demifaces
Acetabulum size Small to moderate Small to Moderate
Moderate As in KSD-VP-1/1 Very large
37
Retroauricular region
Extremely narrow Not well preserved but likely narrow
Very broad Moderate, likely a consequence of body size
Very broad
Superoinferior height of os coxa
Very tall with greatly elongated supra-acetabular terminus
Moderately broad Short and broad As in KSD-VP-1/1 As in KSD-VP-1/1
Posterior superior iliac spine
Little or no expansion of iliac crest at posterior terminus
Not preserved Massive Much more gracile than KSD-VP-1/1, but likely a consequence of body size
As in KSD-VP-1/1
38
Table S4 Ratio of Vertical to Horizontal Tangent Lengths of the Lateral Condyle*
Taxon/Specimen N Mean Ratio Range
H. sapiens 15 47.5 (3.9) 41-53 P. troglodytes 15 65.9 (8.2) 55-82
A.L.333-3 46 A.L. 129-1 48
KSD-VP-1/1 (52)**
*Data from 11. **The lateral condyle has suffered significant damage, though its overall dimensions appear intact. However, the exact location of the meniscal groove must be estimated because of surface damage to the condyle in its most likely location.
39
TABLE S5 ANALYTICAL TRAIT TYPES
Type 1. A trait that differs in two (or more) taxa because its presence and/or expression are downstream consequences of differences in the positional information of its cells and their resultant effects on local pattern formation. Type 1 traits are fixed by directional and/or stabilizing selection because their primary functional features have a direct effect on fitness, and result largely from an interaction between genes expressed during morphogenetic field deployment and the functional biology of their adult product. Such traits are particularly subject to modification by changes in the action and or position of enhancers, silencers, and other elements of cis, trans and mirna regulation.
Example: The superoinferior shortening of the ilium in hominids. Type 2. A trait that is a collateral byproduct of field changes as in Type 1s, but is itself not under direct selection. Type 2 traits differ in two (or more) taxa because of accumulated differences in positional information (as in Type 1's), but have no functional consequences other than their association with the principally selected character. They may, for example, be the consequence of "hitch-hiking" (selective sweeps). Unlike Types 4 and 5, they do represent true field derived pleiotropy. Type 2A The antecedent Type 1 is under direct selection which largely determines its frequency; Type 2B The antecedent Type 1 is not under direct selection (though it may have been so in the past) Examples: Type 2A: Emergence of separate centers of ossification for greater trochanter and femoral head in some mammals--i.e., the principle adaptation being selected is length of the femoral neck17. Type 2B: Degree of expression of the intertrochanteric line of the human femur. This character varies extensively in human populations and clearly has a genotypic basis, but there is no evidence that it is the result of traction force during ontogeny (i.e., it is not a Type 4), or that it has any impact on the function of the iliofemoral ligament18. Type 3. A trait that differs in two taxa because of modification of a systemic growth factor that affects multiple elements, such as an anabolic steroid or systemic expression of cell surface receptors. Examples: Body size and its allometric effects. Allometric shifts probably reflect differential corporal effects of slight changes of systemic control factors during development. Modulations of Growth Hormone or Insulin-like Growth Factor I (or their receptors)19. Specific example: Systemic increase in all elements of the forelimb in Pongo and Hylobates.
40
Type 4. A trait that differs between two (or more) taxa because its presence/absence and/or “grade” are attributable exclusively to phenotypic effects of the interaction of "systemic assembly mechanisms"20 and environmental stimuli. Such traits have no antecedent differences in pattern formation, and therefore have no value in phylogenetic analysis. They are epigenetic and are not pleiotropic. However, they provide significant behavioral information, and are therefore of expository or evidentiary value in interpreting behavior. They often result from habitual behaviors during development. Examples: The bicondylar angle of the femur; depth of the cotylar fossa of the talus. Type 5. Traits arising by the same process as Type 4's (phenotypic plasticity) but which have no reliable diagnostic value with respect to behavior. Such traits are not consistently expressed within species and often show marked variation of expression even within individuals of local populations. Example: differential growth of ear size in response to ambient temperature during ontogeny21. However, such factors can serve as the basis for genetic assimilation.
41
Table S6 Tibia Length Relative to Acetabular Diameter and Other Metrics in KSD-VP-1/1
and in Other Hominoids (N = 25) Used in This Analysis
Character KSD-VP-1/1 A.L. 288-1 H. sapiens P. troglodytes G. gorilla Tibia Maximum Length
355
(227-238) ¶
376.2 (27.3)
254.3 (12.4)
288.6 (32.1)
Tibia Physiological Length
343
363.1 (27.0)
238.0 (12.0)
271.5 (31.2)
Acetabulum Diameter
(49)* 38*** 54.4 (4.4)
40.8 (2.9)
54.3 (6.6)
Femoral Head Diameter
(38)** 28.6 45.7 (4.1)
33.5 (2.3)
45.3 (5.8)
Femoral Length (418-438) ¶ 280 460.4 (26.2)
299.0 (14.6)
347.2 (37.1)
Acetabulum/tibia Ratio
.14 (.16 -.17) .145 (.011)
.161 (.011)
.188 (.012)
*Estimated because of moderate superoinferior crushing **Average of three estimates; for details see text ***The value listed in 22 is 34.7 but represents the inside diameter; the method used here is that described by C.V. Ward23
¶Estimated by crural indexes of 81 and 85
42
Table S7
Metric Parameters of Distal Tibia in Hominoids from DeSilva (5,6, Personal Communication) and this study for KSD-VP-1/1
Taxon
Malleolar Shape Index (AP ÷ ML)/100
Plafond Declination Angle (PDA)
Relative Articular Surface Depth
Metaphyseal Shape Index (AP ÷ ML)/100
N Mean (SD) Range Mean (SD) Range Mean (SD) Range Mean (SD) Range Gorilla 44 68.3 (5.2) 56.7-84.5 105.6 (2.5) 99.3-110.0 12.0 (2.6) 7.9-17.7 63.7 (4.7) 52.7-72.5 Pan 53 72.2 (6.5) 55.7-90.2 102.7 (4.2) 94.5-112.7 12.3 (2.3) 7.6-17.2 69.5 (4.9) 58.6-83.8 H. sapiens 69 54.5 (5.2) 39.0-64.4 91.1 (2.4) 86.1-96.5 16.1 (1.7) 11.5-19.4 82.2 (4.8) 74.6-94.2 A. afarensis 3 59.1 (2.6) 56.1-60.6 90.8 (0.69) 90.0-91.2 16.6 (0.8) 16.0-17.5 85.8 (6.9) 81.0-93.8 A. anamensis 1 52.1 93.1 17.9 84.5 Other early hominids
7-9¶ 56.2 (5.6) 49.6-65.6 91.0 (2.5)* 87.4-94.7* 14.8 (1.6) 10.4-17.8 89.1 (7.6) 80.7-100.3
KSD-VP-1/1 47 100** (20)** 100 ¶Varies: for details see 5,6. *Excludes pathological specimen (probable fibular fracture) KNM-ER-2596. **Impacted by probable sequelae to remote fibular fracture (see text).
43
Table S8 Absolute and Relative Clavicular Length and Circumference in Hominoids
Taxon* Clavicle
Length Mid-Clavicle
Circum. Length/Geometric
Mean Upper Limb
Circum./ Geometric Mean
Upper Limb Pan 128.7 (8.8) 35.8 (2.1) 4.7 (0.34) 1.3 (0.07)
Gorilla 151.2 (20.6) 43.2 (7.9) 4.1 (0.24) 1.2 (0.12) Homo 148.7 (11.3) 39.2 (4.6) 5.2 (0.32) 1.4 (0.12)
KSD-VP-1/1f 156.4** 46 5.3 1.6
*N for each taxon = 25 **Estimated by regression; likely a minimum value
Table S9 Humerus Metrics for Extant Hominoids, KSD-VP-1/1 and A.L. 288-1
Taxon* Distal Articular
Breadth Trochlea Breadth
Humerus Max. Length
P. troglodytes 45.7 (13.0) 27.2 ( 2.3) 306.9 (14.9) G. gorilla 61.0 ( 8.8) 39.2 (6.4) 411.9 (40.1) H. sapiens 44.9 (4.6) 27.3 ( 2.9) 324.8 (20.8) KSD-VP-1/1b 46-48 27 A.L. 288-1 30.1 19.3 235
*N for each taxon = 25
44
Table S10 Ulnar Metrics in Anthropoids*
Taxon Olecranon Orientation
(OO) Mean (S.D.)*
Proximodistal height of olecranon
(OH) Mean (S.D.)*
Keeling Angle (UKN) Mean
(S.D.)*
Trochlear Notch
Orientation (UNO) Mean
(.S.D.)*
Trochlear Notch
Orientation: Proximal Ulna (this
study) N = 15 each taxon**
H. sapiens 110.0 (7.3) 4.2 (1.5) 123 (8) 21.4 (5.5) 27.3 (3.8) P. troglodytes 102.0 (7.2) 3.5 (2.3) 105 (8) 20.5 (7.7) 29.8 (4.9) P. paniscus 105.4 (5.7) 2.0 (2.5) 109 (10) 19.9 (3.8) Gorilla 97.4 (7.0) 1.9 (2.3) 116 (7) 27.3 (5.2) 35.3 (5.6) A.L. 288-1 118 5.1 131 9/10¶ 17 OH-36 122/125*** 12 102 7/10 L40-19 (Omo) 111 5.1 121 2/9.5 A.L. 438-1 125 7.7 131 10.5/13.3 KSD-VP-1/1 7.5 126 10-12 BOU-VP-12/1 127-132 5.5 120 5/8 16 ARA-VP-7/2-C 19# ARA-VP-6/500-051 14#
*Data from 14,15, except where noted. OO: "olecranon orientation measured as the angle between
the axis of olecranon length and the long axis of the bone"; OH: "proximodistal olecranon height measured from the anterior tip of the anconeal process" UKN: Keeling of trochlear notch measured as the angle between the articular processes of the trochlear notch on each side of the keel on the proximal part of the trochlear notch; UNO: Trochlear notch orientation measured as the angle between two lines; one drawn between the two anterior-most points of the anconeal and coronoid processes, and the other between the deepest point on the trochlear articular surface and the center of the ulnar head (15: p. 609).
**UNO values are lower in complete ulnas as a consequence of diaphyseal curvature--compare method described in Figure S27 to Fig 14 in 15.
***Paired values for single individuals are minima and maxima. #Values are estimates (for details see Figure S32 in 16) ¶Values from 14 (i.e., 9) and this study (i.e., 10) were obtained by using methods described in 14,
and were reported in Table S5 of 16. The values plotted in Figure S27 of this study (and in Figure S32 of 16) are by methods described here (and in 16) for evaluation of incomplete ulnas. See also note **.
45
Table S11 Humerus/Femur Ratio in Hominoids
Taxon N Mean H. sapiens 29 71.2 (2.20) P. troglodytes 30 101.7 (3.01) G. gorilla 29 116.4 (3.96) P. pygmaeus 19 130.1 (3.13) H. lar 14 116 (2.06) A.L. 288-1 84
Table S12 Crural Index* in Hominoids (N =50 each taxon)
Taxon Mean H. sapiens .816 (.038) P. troglodytes .847 (.019) G. gorilla .829 (.023) Mean: all three taxa .831
*Tibia maximum length divided by femur maximum length.
46
References
1. Deino A, et al. (2010) 40Ar/39Ar dating, paleomagnetism, and tephrochemistry of Pliocene strata of the hominid-bearing Woranso-Mille area, west-central Afar Rift, Ethiopia. J Hum Evol 58: 111-126.
2. Haile-Selassie Y, et al. (2007) Preliminary geology and paleontology of the Woranso-
Mille, Central Afar, Ethiopia. Anthro Sci 115: 215-222. 3. Gradstein FM, Ogg JG, Smith AG (2004) A Geologic Time Scale (Cambridge University
Press, Cambridge) pp. 1-589.
4. Lovejoy CO (2005) The natural history of human gait and posture 1: spine and pelvis. Gait and Posture 21: 95-112.
5. DeSilva JM (2008) Ph.D. University of Michigan.
6. DeSilva JM (2009) Functional morphology of the ankle and the likelihood of climbing in early hominins. Proc Natl. Acad. Sci. U. S. A. 106: 6567-6572.
7. Wiltse LL (1972) Valgus Deformity of the Ankle: A sequel to the acquired or congenital abnormalities of the fibula. J. Bone Joint Surg Am 54A: 595-606.
8. Simpson SW, et al., (2008) A female Homo erectus pelvis from Gona, Ethiopia. Science
322: 1089-1092.
9. Tague RG, Lovejoy CO (1998) AL 288-1--Lucy or Lucifer: gender confusion in the Pliocene. J Hum Evol 35: 75-94.
10. Robinson JT (1972) Early Hominid Posture and Locomotion (University of Chicago Press, Chicago).
11. Lovejoy CO (2007) The natural history of human gait and posture 3: Knee. Gait and Posture 25: 325-341
12. Jellima LM, Latimer B, Walker A (1993) The rib cage. in: The Nariokotome Homo erectus skeleton (WalkerA, Leakey REF, Eds). (Harvard Univ. Press, Cambridge, Ma.) pp. 294-325.
13. Asfaw B, et al., (1999) Australopithecus garhi: A new species of early hominid from Ethiopia. Science 284: 629-634.
14. Drapeau MS (2004) Functional anatomy of the olecranon process in hominoids and plio-pleistocene hominins. Am J Phys Anthropol 124: 297-314.
47
15. Drapeau MS, et al. (2005) Associated cranial and forelimb remains attributed to Australopithecus afarensis from Hadar, Ethiopia. J Hum Evol 48: 593-642.
16. Lovejoy CO, et al. (2009) Careful climbing in the Miocene: the forelimbs of Ardipithecus ramidus and humans are primitive. Science 326, 70: 70e1.
17. Serrat MA et al. (2007) Variation in mammalian proximal femoral development: comparative analysis of two distinct ossification patterns. J Anat 210: 249-258.
18. Lovejoy CO, Heiple, KG (1971) The proximal femoral anatomy of Australopithecus. Nature 235:175-176,
19. Serrat MA, Lovejoy CO, King D (2007) Age- and site-specific decline in insulin-like growth factor-I receptor expression is correlated with differential growth plate activity in the mouse hindlimb. Anat. Rec. (Hoboken. ) 290: 375-381.
20. Lovejoy CO, Cohn MJ, White TD (1999) Morphological analysis of the mammalian postcranium: A developmental perspective. Proc. Nat. Acad. Sci. 96: 13247-13252.
21. Serrat MA, King D, Lovejoy CO (2008) Temperature regulates limb length in homeotherms by directly modulating cartilage growth. Proc Nat Acad Sci 105: 19348-19353.
22. Johanson DC, et al. (1982) Morphology of the Pliocene partial hominid skeleton (A.L. 288-1) from the Hadar Formation, Ethiopia. Am J Phys Anthro 57: 403-451.
23. Ward CV (1991) Ph.D. Johns Hopkins.